Which Half of Promotions Work?

‘Half of all promotions are wasted, but which half?’, to paraphrase the famous advertising saying attributed to 19th century industrialist John Wanamaker.

The need for trade promotion post event analysis should be self-evident in order to adjust trade spend and planned future promotions, and to minimize trade overspends. But companies’ ability to consistently analyze promotion results in a timely and comprehensive manner remains patchy.

This may be due to a lack of organizational capability and tools for analysis, not knowing what’s most important to measure, the variety and complexity of data sources required for thorough post event analysis (PEA), or all of the above.

What good PEA looks like?

Individual Promotions

Whilst comprehensive understanding of an individual promotion may be the goal, you don’t have to boil the ocean to achieve it. There are 5 fundamental questions all CPG companies should be able to answer about an individual promotion’s performance:

1. Did the promotion you expected to run, actually run?
2. Did you get the promotional ACV (annual contract value) you were expecting?
3. Were your projected promotional volumes close to the actual volumes?
4. Are there competitive products that perform well with us or vice versa?
5. Was the promotion profitable?

At a minimum this involves analyzing an individual promotion’s performance versus forecast; ROI, margin and incremental profit evaluations, competitive and consumer event overlays – often involving external data source integrations. 

Additional sources of insight also come from understanding and integrating the following:

• Capturing true promotional uplifts as increments including promotion mechanics and objectives (penetration, frequency, AWOP), pantry loading, forward buying, price elasticity, duration; frequency and depth; timing/seasonality and pre and post promotion volume dips
• Cannibalization of own products as well as competitor products 
• ‘Store back’ attributes from the point of purchase including product availability, in-store compliance, regional and geographic considerations, and retailer feedback
• Competitor promotional program calendar (including conflicts with your promotions) and performance.

Further asking the ‘why’ and ‘why not’ questions allows for future planning. The ‘would we do this again, in the same way?’ question. Rolling up PEA at a customer and category level also enables development of a roadmap for future successful promotional activities.

Using PEA to set norms and benchmarks

Beyond understanding individual promotion performance, PEA can be used for planning future promotions – trade promotion optimization (TPO).  Something which enables you to see what works and doesn’t over time, by promotion and by retailer, in order to create ‘norms’.

You can look at trends by individual category, quantify sales anomalies and spikes in promotional periods, and compare an individual promotion’s ROI to prior promotions. You can determine average or benchmark promotional performance and uplifts by mechanic.

All of this allows you to set uplift predictions for future promotions including cannibalization effects, as well as likely product sell in and sell out for demand planning. By using PEA to understand norms, you can use it to set promotion guidelines and guardrails.

For instance, Welch’s Foods were manually evaluating promotions using spreadsheets, which aside from non-repeatable practices and lost time, and high risk of repeating poor performing promotions and unnecessarily deep discounting, also meant siloed processes and difficulty in applying consistent sales planning guardrails. By implementing a TPO system with consistent PEA capabilities, trade investment ROI increased 16%+ by identified and replacing underperforming promotional tactics. Welch’s salespeople could see at the touch of a button, the effectiveness and profitability of not only the previous year’s promotion but the predicted outcomes of this year’s promotional calendar.

Easy PEA-sy with harmonized data 

True PEA gives you the ability to utilize multiple sources of data in one view. That means POS/scan, syndicated data, internal shipments and spending, pricing, competitive activities, store back data, loyalty program data, and all the other sources we listed above. And these come from multiple sources internally, from retailers, data suppliers and other sources of intelligence.

Using a TPM/O system not only centralizes the multiple data sources required for PEA but standardizes – ‘harmonizes’ – the language and metrics from the various data sources so that a holistic view is provided with minimal inaccuracies and anomalies.

Data Harmonization has a number of steps. After acquiring the various identified internal and external data, all of the data is ‘ingested’ into a central system where it is cleansed and combined. It is then harmonized based on your business and product hierarchy. Language is smoothed to remove discrepancies and outliers. For example, an item that had a different description in three different data sources will now have a consistent description and performance from all three sources that can be viewed in one place. Following harmonization, custom market metrics are layered in, and then analysis begins in order to plan predictively and predictive planning, perform ‘what if’ scenario modelling, and to optimize calendars.

PEA using a TPO system enables companies to understand individual, past and future promotional performance and to optimize spending and investment by determining the tactics behind good, and bad, performance. It helps you avoid the definition of promotional insanity – doing the same things repeatedly but expecting different results – by harmonizing data from multiple sources and comparing like for like using the same language.

In other words, finding out which ‘half’ of your promotions are working, and improving this over time.

To learn more about using data harmonization for Trade Promotion Post Event Analysis, or to schedule a demo, visit: https://blacksmithapplications.com/ and www.exceedra.com